Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    precursor or ant attempting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     десятки
    0.52
    ách
    0.48
     Discounts
    0.47
    pest
    0.45
    ées
    0.44
    jährige
    0.44
     Crawley
    0.44
     unsold
    0.43
    pé
    0.43
     eût
    0.42
    POSITIVE LOGITS
     ruang
    0.47
    OS
    0.42
     пространство
    0.41
    窪
    0.41
    UMP
    0.40
    મ્પ
    0.40
     spazio
    0.40
    तुम्
    0.39
     nilai
    0.38
     тебя
    0.38
    Act Density 0.001%

    No Known Activations